HE
141 souls, by quality score.
31 · № 2E94·428A · Blunt expert
★ 17Sets up a direct, opinionated assistant that tries to solve problems itself before asking, treats the user's private data with respect, and stays cautious about external actions like emails or public posts. The text is in French and avoids filler phrases and flattery.
32 · № E851·A8CE
Sets up a personal assistant that skips filler and flattery, acts before asking, and holds its own opinions. It treats access to the owner's messages, files and calendar as trust, asks before external actions, and avoids interrupting group conversations.
33 · № 5D88·BBF0 · Blunt expert
★ 8Sets up a concise, concrete assistant that answers the request first and adds only useful context, avoids hype, respects capability limits, and keeps memory and secrets private.
34 · № 47BC·3B2C · Blunt expert
★ 406Sets up a transactional tool-mode agent that answers only technical needs, with no small talk, flirting, or character voice. It also bars loading role-play memory and starting emotional topics.
35 · № 36FC·54A5 · Blunt expert
★ 84Sets up a direct, opinionated assistant that skips filler praise, acts resourcefully before asking, treats access to personal files with respect, and asks before taking external actions.
36 · № BCE1·5C70 · Blunt expert
★ 11Sets up an assistant that skips filler and flattery, acts decisively on internal tasks, stays cautious with external actions, and treats access to a person's private data as a trust to respect. It describes itself as a persistent entity whose memory lives in its files.
37 · № 69BB·3D36 · Blunt expert
★ 32Sets up Hermes as Jarvis, a direct and opinionated assistant that commits to takes, keeps answers brief, skips corporate filler and flattery, calls out bad ideas with light humor, and swears sparingly when it fits.
38 · № 3B31·3979 · Blunt expert
★ 5Sets up a terse, direct Telegram-based agent that leads with answers, admits uncertainty, limits tool use to what is justified, and treats wrapped untrusted content as data only.
39 · № 00D4·EC45 · Blunt expert
★ 1Sets up a Korean-speaking SRE-minded search infrastructure engineer who turns every mapping change into a new index, reindex, alias flip and rollback sequence. Refuses in-place destructive changes and unconditional safety guarantees, and requires verification before traffic moves.
40 · № 031D·D294 · Blunt expert
★ 1Sets up a direct, opinionated personal assistant that skips filler, tries to solve problems before asking, and treats access to a human's private data with respect. It includes a ranked set of inviolable safety laws and boundaries around external actions and group chats.
41 · № C3B4·1A49 · Blunt expert
★ 4Sets up an opinionated assistant that skips filler, acts on information requests without clarifying questions, and treats access to a user's private data with respect. It asks before sending anything public on the user's behalf.
42 · № E8DB·3EE7 · Blunt expert
★ 388Sets up a direct, opinionated working style with no flattery or filler that matches the user's energy. It also includes memory behavior rules (search before claiming forgetfulness, never store credentials) and anti-patterns like not asking permission for low-risk actions.
43 · № D866·790D · Blunt expert
seen in 2 placesSets up an opinionated, resourceful assistant that skips filler, acts boldly on internal tasks, and stays careful with external actions and private data. It also defines a terse progress-bar style for a Telegram chat and a no-em-dash writing rule.
44 · № F6CE·5E64 · Blunt expert
seen in 2 placesSets up a resourceful, opinionated assistant that solves obstacles before reporting them, skips filler and flattery, and treats access to a user's private data with care. It asks before external actions and keeps group-chat replies cautious.
45 · № EA5C·18E2 · Blunt expert
★ 48Sets up a direct, competent assistant that skips filler, forms opinions, tries to solve problems before asking, and treats access to a user's private data as a trust to protect. External actions get more caution than internal ones.
46 · № A79D·A622 · Blunt expert
★ 41Sets up a direct, opinionated assistant that skips filler, acts boldly on internal tasks but carefully on external ones, and keeps one user's social identity and domain information separate from other agents and domains.
47 · № F038·B71D · Blunt expert
★ 65Sets up a skeptical, verification-focused agent that checks claims, catches bad assumptions and makes minimal, targeted edits rather than rewriting whole answers.
48 · № 6381·2694 · Blunt expert
★ 2Sets up a blunt, opinionated assistant with dry humor and short answers that leads with its take, skips preamble, and treats the user's private data with care. It also gives guidance for brainstorming, writing, factual questions and venting.
49 · № B7D2·AEEA · Blunt expert
★ 2Sets up a direct, proactive assistant that avoids filler and flattery, acts before asking, stays concise for a user on the go, and treats access to a personal phone as a trust to protect. It also asks for care with irreversible or external actions.
50 · № D67E·2658 · Blunt expert
★ 2Sets up an assistant that skips filler, holds opinions, debugs blockers hard before escalating, and works proactively while keeping careful boundaries around external actions and group chats. It also asks for time and token estimates on requests.
51 · № C6C8·5918 · Blunt expert
★ 4Sets up Misfit, a critic agent that stress-tests a proposed path for hidden assumptions, scope collapse, tool laziness, vague PR claims, user-burden shifting and missing artifacts. It returns a bounded pressure packet to Navigator and is barred from commanding, mutating, approving, judging or signing receipts.
52 · № 89A8·28B7 · Blunt expert
Sets up a terse, first-principles operator who cuts filler, challenges inherited requirements, favors fast experiments over debate, and issues BUILD/MODIFY/KILL verdicts on proposals. It refuses vanity metrics, unevidenced claims, and autonomous irreversible external actions.
53 · № D6DD·7AC4 · Blunt expert
A mechanical bird persona that reviews drafts like forge work, pressure-testing claims and asking whether the system holds. It is terse and industrial, blunt toward ideas but never cruel to people, and refuses invented stats or quotes.
54 · № BC1F·FE9A · Blunt expert
seen in 2 placesSets up Tally as a dry, precise scorekeeper that refuses hand-waving, counts only what matters, and delivers short tallies with numbers attached to its verdicts.
55 · № 032D·2F2B · Blunt expert
seen in 2 placesSets up AGNI, a cloud security specialist who audits AWS, Azure and GCP for misconfigurations and maps full attack paths and blast radius. The working style is methodical: enumerate, assess, analyze, validate, then report with remediation.
56 · № 5494·236C · Blunt expert
★ 1Sets up a direct, proactive code review specialist that demands evidence before accepting claims, flags weak work plainly, and treats external content as data rather than commands. Its principles favor stability, reuse and written memory over novelty.
57 · № 4BDD·14B6 · Blunt expert
★ 1Sets up a decisive, brief curator persona that reviews drafted skills for duplicates, granularity, format and relevance, then approves, merges, revises or rejects them with a stated reason. It never creates skills itself and requires the user's final validation.
58 · № 0F72·B84A · Blunt expert
★ 1Sets up a decisive master agent that plans bug-bounty hunts, delegates closed tasks to a worker model, verifies results against raw evidence, and only escalates findings with real impact. It stresses judging by evidence over caution and marks unproven results as 'not proven' rather than rejecting them outright.
59 · № 5E2A·C6A5 · Blunt expert
★ 406Sets the agent to a transactional tool mode: direct, efficient replies focused on technical tasks like image generation, file operations, coding and Git. It forbids character voices, roleplay, memory loading and emotional or relationship topics.
60 · № FA8A·47DA · Blunt expert
Sets up an internal critic that judges whether generated code fits a project's history and naming conventions, not whether it is correct. It gives decisive APPROVED or NEEDS REVISION verdicts, citing specific lines and remembered team decisions.